zhifu gao
zhifu gao
Sorry, the funasr runtime currently does not support this`damo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn` model.
\>=16GB
The feature of ITN is now in progress of testing, maybe you could test it on: https://101.37.77.25:1333/static/index.html
Sorry, we do not have such plans. You can pre-process the audio by cutting it according to the segment information before proceeding with model training.
I have tested it with the ct-transformer large version, the result is expected. Maybe you could try it: https://github.com/alibaba-damo-academy/FunASR/tree/main/egs_modelscope/punctuation/punc_ct-transformer_cn-en-common-vocab471067-large
Please raise issues ref to https://github.com/alibaba-damo-academy/FunASR/issues/1073
> 目前的代码（master head）onnx 在 windows 下无法编译（Onnx runtime 1.15.1），希望支持一下，自己通过修改代码解决了一下，因为处理的比较乱，没好意思提交 PR： > > ### 支持 windows 修改： > #### 1. Onnx session 构造函数签名宽字符问题. > 例如： > > ```c++ > std::make_unique(env_, punc_model.c_str(), session_options);...
The FunASR community software package for Windows has been adapted and now supports both offline file transcription and real-time dictation). We welcome everyone to try it out and provide feedback....
Please raise a issue following https://github.com/alibaba-damo-academy/FunASR/issues/1073
Please raise a issue as https://github.com/alibaba-damo-academy/FunASR/issues/1073